Instructions to use OdyssAI-Pro/microduck-1-2-3-quack with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Microduck
How to use OdyssAI-Pro/microduck-1-2-3-quack with Microduck:
# Replace SLOT with the slot specified in the model card (walk, stand, sitstand, ground_pick, kick_left, kick_right, roulade). sudo robotctl policy load SLOT OdyssAI-Pro/microduck-1-2-3-quack
- Notebooks
- Google Colab
- Kaggle
1-2-3-quack
1, 2, 3, Quack! A walker turns around, the choir freezes. Choreography only: Pollen's Got a Talent reference policy, unchanged, for the 5 singers; the conductor is dashi/microduck-backroll and the walker RemiFabre/microduck-rough-walk-e.
A perpetual policy for the microduck (61-D observation, 14 actions, 50 Hz). Runs until told otherwise โ a gait, loaded into a policy slot.
Run it on a robot
sudo robotctl policy load <slot> Expertnocode/microduck-1-2-3-quack
The observation normalizer is baked into policy.onnx; feed raw observations.
manifest.json follows schema 2 of the microduck policy manifest (docs/policy-manifest.md in the daemon repo).
A remix of pollen-robotics/microduck-got-a-talent-reference.
Training
Not trained here. policy.onnx is pollen-robotics/microduck-got-a-talent-reference, unchanged. The work in this repo is the choreography in timeline.json, which also borrows dashi/microduck-backroll (the conductor) and RemiFabre/microduck-rough-walk-e (the walker).
- Downloads last month
- 18